top of page

If You Need to Think, You’re Doing It Wrong

A month ago, my wife and I were in Barcelona for a short vacation.

In one of the days, we entered a small place called Wok to Walk for dinner. The concept is simple: You build your dish by choosing your base (the type of noodles or rice), then your favorites (chicken, beef, tofu and so on), the toppings and the sauce.

While waiting for the food, I started to notice the process in which the dishes are being made. Here’s how it works:

Step 1:

The cashier takes your order and puts the ingredients in a bowl. She then takes the paper with the dish details, sticks it on a take away box, puts the bowl on the box, and slides the box to the end of the dishes queue.


Step 2:

The cook grabs the box from the queue and looks at the paper to see what sauce and other stuff he needs to add to the dish.


Step 3:

The cook throws the ingredients to the wok, adds the necessary sauce and cooks the dish.


Step 4:

When the dish is ready, the cook throws it back to the box.


Step 5:

The cook slides the box to the “ready dishes queue”, where the cashier will grab it and give it to the hungry client (the cook will sometimes skip this step and give the dish to the client himself).


The beauty in the process and the reason for its efficiency is that no one along the process has to think.

And when you don’t have to think, you can do advanced things like cooking two dishes at once:

Well, you may say, “I’m in a position where getting paid to think”. That’s true, but are there places in your job where thinking is not needed and hurts your efficiency?

Thinking ABOUT the process is more beneficial than thinking DURING the execution of the process.

In an interview on the Away From the Keyboard podcast, Brent Ozar said that “if you’re typing DMV contents by hand in the year 2015, you officially suck”, referring to the fact that having a process and using prewritten scripts is more important than remembering which DMV holds which column and typing queries yourself.

To continue what Brent said, analyzing an execution plan usually requires thinking, but the way to get this execution plan can mostly be streamlined into a process so you don’t need to reinvent the wheel over and over. Same goes for understanding why a server is slow, finding out if you have a storage problem, installing a new server, adding a replication subscriber, and so on. Some (not all) of those tasks require some thinking at the end of the process, but since a process is in place, you can focus your thinking where it really matters, and not on actions that can be automated or semi-automated.

In other words, if you perform a repeating task, it can most likely be turned into a process. 

A process includes:

  1. A series of steps (1 or more)

  2. The tools to execute each step (a predefined script/your brain)

  3. The order in which you execute each step

When we perform our Data Review process, we follow a checklist we wrote once with scripts we defined for each step in the checklist. We then deliver the client a report detailing the most important things he needs to fix in his system. The template of this report is also predefined so we don’t have to think about the template and focus our thinking cycles where it really matters – analyzing the data.

When we produce the SQL Server Radio podcast, we also have a process:

  1. Meet on the second Monday of the month to record 4 shows

  2. During the recording, write notes about the items and subjects we talk about, and about hard faults to edit out

  3. Write the show name and planned publishing dates in the editorial calendar (a fancy name for an Excel file)

  4. Record promos for all 4 shows

  5. Towards publishing date:

  6. Edit out needed sections according to step 2

  7. Add promo, intro music and outro music (which I have as files, so I just need to copy-paste them)

  8. Produce mp3 file

  9. Upload file to audio hosting

  10. Write show notes

  11. Schedule for publishing

The process sounds long and complicated, but we are efficient executing it, and most of the thinking was done when putting it in place and optimizing it, not while executing it. For example, by writing the items and hard faults during the recording, I don’t have to hear the recording afterwards in order to find the items and things to edit out.

What are the tasks you perform over and over? Think about a process for them, build it, and use your thinking cycles where they really matter.

Thinking ABOUT the process is more beneficial than thinking DURING the execution of the process.




Get New posts delivered straight to your inbox

Thank you for subscribing!

bottom of page